Hearing in the Representational Sense

As for hearing in the representational sense, the case is more complicated.

In an ordinary auditory experience, nothing appears to be at our eardrums. Phenom-enologically speaking, we hear only something at a distance from us, and this thing is the event source. It is therefore far from obvious that the constituent fluctuations causing our auditory experiences are represented. However, there are occasions where we can identify in our auditory experience phenomenal features which do not correspond to the physical features of the event source.

35 I glossed over the ~0.05 seconds processing time in the brain (Nevid, 2016, p. 100).

Consider this case. Suppose an orchestra is playing on the stage and you are allowed to walk around in the auditorium to choose the seat with the best sound.

When you walk around, you hear the orchestra as staying at the same place, while at the same time you are aware of how your auditory experience varies systemati-cally with your changing location. You can thus realise that your experience repre-sents some features distinct from those of the performance, and it thereby also rep-resents the bearer of those features. Since these features vary with your location, they are tied to your location rather than to the performance. However, these fea-tures do not appear to be at your location, as nothing appears to be so located. Rather, their location can only be learnt through reflecting on your experience. My sugges-tion is that such features are features of the constituent fluctuasugges-tions next to your eardrums, and hence they are local features of the compression wave produced by the performance.

Your auditory experience in the above situation is both an experience of the performance and an experience of the compression wave. It is an experience in vir-tue of which two facts of representation obtain: the fact that it represents a perfor-mance and the fact that it represents a compression wave. The second fact should not be taken as having any implication on what it is like to hear compression waves.

Indeed, at least in ordinary cases, auditory experiences always appear to be about event sources. Even in the above example of walking around in the auditorium, it would not be phenomenologically accurate to say that you hear something distinct from the performance. There is one sense in which the performance appears softer when you walk over to the back of the auditorium; there is another sense in which the performance appears equally loud despite your changing location. We may call the first “perspectival loudness” and the second “intrinsic loudness”. Phenomeno-logically speaking, both perspectival loudness and intrinsic loudness appear to be features of the performance. Neither of them appears to belong to something distinct from the performance. It is only through reflection that we realise that the perspec-tival loudness should not be attributed to the performance but rather be treated as the local loudness of the compression wave.

Notice that in the representational sense of hearing, we hear constituent fluc-tuations rather than total flucfluc-tuations. This is because our auditory system analyses the detected total fluctuations into smaller units which in ideal cases correspond to the constituent fluctuations contained in the total fluctuations.

A likely alternative account is to take the perspectival loudness as really a property of the performance on the stage. There is no need to deny that the two constituent fluctuations next to your eardrums immediately cause and thus explain the phenomenal quality of your auditory experience. However, this does not imply that your experience represents such local events—a representation need not repre-sent its causes.

It can be expected that distal theorists would prefer this alternative explana-tion. This is because they can avoid turning sounds—distal entities at the sound sources for them—into indirect objects of our auditory experiences. Even if they reject sonicism, it is unlikely that they would accept that sounds as distal entities are heard in the representational sense via hearing something in the medium. How-ever, since I am now merely explicating what it means to hear compression waves in my wave theory, I need not reject this alternative explanation here.

Since in terms of phenomenology it seems we only hear event sources, there is a straightforward answer to the question of what it is like to hear compression waves: it is like hearing event sources. It is thus tempting to describe compression waves as perceptually “transparent”: if you attempt to attend to a compression wave, you will end up attending to the event source and its features. However, there is a worry: if compression waves are not something we can single out in auditory expe-riences, it is then quite perplexing how auditory experiences can in any sense be counted as experiences of compression waves. Since sounds are compression waves in wave theory, this threatens the status of sounds as objects of auditory experiences.

We may respond by questioning the accuracy of describing compression waves as perceptually transparent. Consider the case of visually transparent objects.

The pane of glass I see through does not share the visual properties of the object behind it. The object is, say, red, but the glass is colourless. Perhaps the object is a bright light source, but the glass is not bright. In the right situation, the pane of glass is invisible. However, this does not seem to be the case for compression waves. As I said in §6.1.3, sound sources inherit the auditory properties of their sounds. Also, saying that compression waves are inaudible may not sound right even to those who deny that we experience them.

The case of compression waves appears to be more like the case of photos, where the representations and the represented objects share visual properties. As Walton (1984, p. 252) says, photos are transparent but not invisible: we see them

while seeing objects in them. Shall we then understand experiences of compression waves as analogous to experiences of photos? Doing so would then be treating com-pression waves as representations of their sources. There is a serious difficulty for this suggestion. It seems our ability to see objects in photos is at least partly ex-plained by our ability to see those objects in real life without the causal mediation of photos. However, it is impossible to hear sound sources without the causal me-diation of compression waves. It is therefore unexplained how we can hear event sources through hearing compression waves. Moreover, seeing an object in a photo is an indirect way of seeing it, and the indirectness is explicit in the experience. In contrast, hearing sound sources is not explicitly indirect—it is phenomenologically direct.

Visually transparent objects—visible or not—do not seem to be a good model for us to conceive of our experiences of compression waves. Therefore, it would probably be better to understand experiences of compression waves in some other sense. The analogy between sound and light introduced in §6.3.3 offers a new suggestion. Since sounds and sound sources are analogous to light and light sources, perhaps we can understand experiences of sounds and their sources by comparing them with experiences of light and light sources.

Philosophers typically deny that we visually experience light when we look at reflective objects. Hilbert (2005, pp. 150-151), for example, says that we see how an object is illuminated but not the illumination itself. However, the case of looking directly at a light source is sometimes treated as an exception to this claim (e.g.

Casati, 2018, p. 167). Let us then accept that light is visible at least when we look at a light source. Do we see light as something distinct from the light source? Phe-nomenologically speaking, this is not the case. Rather, the light seems to be insep-arable from the light source. It does not look like properties such as colour in ap-pearing to be bound to an object. Nor does it appear to occupy a location different from that of the light source. Indeed, there does not seem to be any difference be-tween seeing the light and seeing the light source.

The case is like what I have said about hearing a sound and hearing a sound source. The identification problem of auditory objects introduced back in §2.1.3 may arise from the experiential indistinguishability between hearing a sound and hearing a sound source. I propose that our experiences of waves and wave sources should be treated as a sui generis kind of perceptual phenomena which should be

studied in their own right. For the moment, I can only discuss an observation and provide what I take to be a plausible preliminary explanation.

I shall now focus back on the case of sounds and sound sources, but I intend my claims to be transferable to the case of light and light sources. A crucial idea is that compression waves are the perceptual means by which we hear event sources.

A compression wave carries information about its event source in virtue of the sys-tematic relation between its local acoustic properties and the acoustic properties of the event source. Our auditory system takes advantage of this relation to retrieve information about the event source. This information retrieval process operates in a way which generates auditory experiences with certain auditory phenomenal properties. Such phenomenal properties represent both the local acoustic properties of compression waves and the intrinsic acoustic properties of event sources.

This qualitative aspect does not exhaust the content of auditory experiences:

the locations of event sources are also represented. When interpreted with certain assumptions by our auditory system, constituent fluctuations can provide infor-mation about the distance between an event source and a perceiver in virtue of their spectra. To locate an event source, however, the auditory system still needs to de-termine the direction from which the compression wave comes.

Directional hearing partly relies on information provided by the interaural time difference (ITD) and the interaural level difference (ILD) between the stimuli of the two ears. Both ITD and ILD are results of the two different paths linking the two ears to the sound source. The two paths may or may not have equal length, depending on the location of the sound source relative to the perceiver on the hori-zontal plane. A compression wave enters the ear closer to the sound source first, resulting in an ITD between the arrivals of the compression wave at the two ears.

Also, on the path between the sound source and the ear further away from it, there is an obstacle—the head. The compression wave thus needs to diffract around it before getting into that further ear, and this leads to a drop in the amplitude and thereby an ILD between the constituent fluctuations at the two eardrums.

Strictly speaking, ITD and ILD are not properties of the constituent fluctu-ations but quantities generated in the hearing process: they are relational properties between the two stimulation events rather than properties of the stimulants them-selves.

In the resulting auditory experiences, such locational information is inte-grated with the qualitative information. Indeed, in ordinary auditory experiences, the qualitative aspect and the locational aspect seem to be equally salient. The au-ditory object thus appears to be an individual which has certain auau-ditory properties and is located at a certain distal location. Since this auditory object appears to have both the qualitative and the locational aspects, it is more phenomenologically accu-rate to characterise an auditory experience as like hearing an event source rather than constituent fluctuations which do not have the represented spatial features.

We can now see a disparity between the spatiality and temporality of how compression waves are represented. Compression waves can be temporally repre-sented because their local features at the eardrums are accessible by the auditory system. The same phenomenal properties of an auditory experience, such as phe-nomenal duration, pitch, loudness, timbre, etc., represent both the perspectival du-ration, frequency, amplitude, and spectrum of the event source and the local dura-tion, frequency, amplitude, and spectrum of the compression wave.

In contrast, compression waves are not spatially represented. Again, our perceptual access to compression waves are limited to their local features at the eardrums. The auditory system has no mechanism to pick out the locational infor-mation about the compression waves. The only locational inforinfor-mation retrievable from the stimulants concerns the location of the event source instead. Therefore, compression waves are only temporally but not spatially represented.

In ordinary cases, the qualitative and the locational aspects seem to be equally salient in our auditory experiences, and hence event sources are the foci of our auditory attention. This does not mean that we cannot focus on compression waves if we want. When I attend to the perspectival auditory properties of an event source in my auditory experience, I implicitly also attend to the local auditory prop-erties of its compression wave at my eardrums, because they are represented by the same phenomenal auditory properties to which I am attending. However, since the phenomenal location only represents the location of the event source, I can attend specifically to the event source if I focus on its location. In contrast, I suspect that it is impossible to attend only to the compression wave, because our auditory expe-riences do not have any phenomenal properties which represent features of com-pression waves only.

One minor worry here is that there are two constituent fluctuations of the same compression wave causing an auditory experience, but they are not experi-enced separately. It might then be questioned whether they are really experiexperi-enced at all. I believe the most plausible reply to this worry is that the resolution of our auditory experiences is not high enough to represent the difference between the two local fluctuations. Our auditory system is sensitive to the tiny differences and makes use of them, for example, in the localisation of sound sources. However, such dif-ferences are not reflected in the auditory properties experienced. In most situations, hearing something with one ear covered would not have any noticeable difference in auditory properties from hearing the same thing with the other ear covered. That said, I should note that in some cases, the difference becomes noticeable if it is large enough. The most obvious case is hearing music with only one earphone on. It is obvious in such an experience that the music and its qualities are represented only on one side. To this extent, we can hear constituent fluctuations individually.

My solution to the unification problem is that, although there seems to be only one thing present in my auditory experience, this experience actually repre-sents both the local parts of a compression wave and its event source. A sound, identified as a compression wave, is experienced not in its entity but only partially.

However, we cannot experience the most essential feature of it—i.e. the propaga-tion of pressure fluctuapropaga-tion. There is no phenomenological difference between ex-periencing a sound and exex-periencing its source, since they are indeed the same act of auditory experiencing. Hearing a sound is just like hearing its event source.

Equivalently, hearing an event source is just like hearing its sound. However, we can realise through reflection that two facts of hearing, or we may say, two facts of auditory experience, obtain in virtue of this experience: the fact that it represents two constituent fluctuations and the fact that it represents an event source.

6.6.3 Sonicism

In Chapter 2, I argued that sonicism cannot serve as a theoretically neutral ground for any theory of sound. Nonetheless, it may still turn out to be true. I pro-ceed to examine it in light of my wave theory in this subsection.

When we focus on the case of sound sources, sonicism is obviously true in the causal sense. Sounds are compression waves, and our ears are immediately stim-ulated by local parts of compression waves. Therefore, sound sources can produce

auditory experiences only via the causal mediation of compression waves. It is, however, unclear whether this can be generalised to all non-sound entities.

Consider the case of bone conduction. Bone conduction involves multiple mechanisms, one of which is of particular interest to us here: inertial bone conduc-tion (Gelfand, 2018, pp. 79-80). The human skull responds to the vibraconduc-tion directly applied to it in different ways depending on the frequency of the vibration. At lower frequency (e.g. 200 Hz), it moves as a whole—i.e. the entire skull moves in the same direction. However, the ossicles are suspended in the middle ear like pendu-lums. Their inertia stops them from moving with the skull and results in relative movements between them. Such movements are basically the same as those which follow the vibration of the eardrum in the case of air conduction, and so the inner ear is stimulated in the same way.

What makes this case interesting is that the whole process of inertial bone conduction can involve no compression wave. The real human skull can be com-pressed, but even if it is completely rigid, nothing essential would be affected in the process of this kind of bone conduction. So, if a vibrator is applied directly to the side of such a completely rigid skull, nothing is compressed and hence this is no compression wave. According to my view, there is no sound, although the subject can hear the signal generated by the vibrator. Therefore, the resulting auditory ex-perience is not caused by the vibrator (a non-sound entity) via a sound.

The vibrator is, in fact, a sound source, as its vibration in air generates a compression wave. However, in the current case, it is not heard in virtue of its sound, so it is not heard qua sound source. If the vibrator is not surrounded by an elastic medium, it cannot produce any compression wave and hence it is not a sound source.

Nonetheless, a subject can still hear it through bone conduction if it is applied to her head directly.

I do not know what sonicists would say regarding inertial bone conduction.

One option is to say that although the vibration of the vibrator produces an auditory experience, this is not the right kind of causal process to qualify the vibrator as being heard. The experience is, accordingly, a hallucination. This option seems to be no more than an ad hoc defence of sonicism.

Another option would be to give up sonicism entirely. This is, in my opinion, throwing the baby out with the bathwater. As I admitted earlier in this subsection,

Another option would be to give up sonicism entirely. This is, in my opinion, throwing the baby out with the bathwater. As I admitted earlier in this subsection,